Ok, here\'s a summary of what happens from everything I\'ve read
1. Loader.bin boots the computer and then runs diskem1x.bin.
2. diskem1x.bin looks for diskemu.cmd to put up your menu choices for you to choose what to install.
3. Choosing a menu choice runs that specific *sect.dat. This file must point to the specific boot floppy directory of what system your trying to install. (ie. ASRV, WPRO, WSRV).
4. Setupldr.bin (in that specific boot floppy directory) now runs. It must point to the boot floppy directory that itself is in.
5. Ntdetect now runs in that same directory. Ntdetect reads txtsetup.sif for the required info to start the install.
Now everything should install properly if everything I\'ve read is true.
All the above is compiled from this thread, Oz\'s work, Bart\'s site, the Chinese website with a similar procedure (after I translated it, I think Space also posted a translated version in this thread). Thanks, everyone!! I\'m still trying to understand all this in an orderly fashion.
Please feel free to correct my summary if anyone notices errors. I have one more important question --
Why does this procedure use the floppy bootfiles/directories if the original Win2k CD doesn\'t. Since the original CD has a lot of the files (mentioned above) pointing to the I386 folder, why can\'t we just substitute \"I386\" in those files with our new folders ie. (UENGLISHPRO or UENGLISHADV_SER or UENGLISHWSRV

Tyee